Screening coronary artery disease with computed tomography angiogram should limit normal invasive coronary angiogram, regardless of pretest probability.

BACKGOUND: Performing functional testing (FT) or a computed tomography angiogram (CCTA) before invasive coronary angiogram (ICA) is recommended for coronary artery disease (CAD). We aimed to evaluate, in a real life setting, the rate of strictly normal ICA following a positive noninvasive test result. METHODS: We included all patients who underwent an ICA with a prior positive FT or CCTA. Patients were categorized in 5 subgroups, according to pretest probability (PTP) of having a CAD. Main results of ICA were defined as normal ICA, nonobstructive CAD (non-oCAD), and obstructive CAD (oCAD). RESULTS: For 4,952 patients who underwent ICA following either a positive FT (3276, 66.2%) or CCTA (1676, 33.8%) result, the PTP was (1) low (<15%; n = 968, 19.5%), (2) lower intermediate (15%-35%; n = 1336, 27.0%), (3) higher intermediate (35%-50%; n = 806, 16.3%), (4) high (50%-65%; n = 806, 17.7%), and (5) very high (> 65%; n = 965, 19.5%). ICA showed no CAD (819 patients, 16.5%), non-oCAD (1,193 patients, 24.1%), or oCAD (2940 patients, 59.4%). Without considering the PTP values, CCTA compared to FT showed less frequently normal ICA (7% vs 16.5%), and more frequently CAD (non-oCAD 27.9% vs 22.2%; oCAD 65.1% vs 56.4%) (all P < .0001). When we considered the different PTP values, CCTA always showed lower rates of normal ICA than the FT. In low- and lower intermediate-risk patients, CCTA detected more frequently oCAD compared to FT (P < .001). CONCLUSIONS: CCTA is a better alternative than FT to limit unnecessary ICA regardless of PTP value, without missing abnormal ICA.